Fetching the paper…

PreCurious: How Innocent Pre-Trained Language Models Turn into Privacy Traps · Around